Dell Inspiron N5110: System time does not advance when unit is off.

[David Farber]

This Dell Inspiron N5110, which has Windows 7 Home Premium installed,
was given to me because upon startup it would BSOD with a
REGISTRY_ERROR. That was resolved with a system restore but I noticed
another problem which has me puzzled. The system time does not advance
when the unit is turned off. I went into the BIOS setup and noticed that
within the time configuration, the seconds were not ticking. Once the
machine is booted into Windows, the automatic time synchronization will
eventually set it to the correct time and it stays that way until it's
powered off. The CMOS battery measures above 3 volts in circuit and all
the other BIOS settings are saved when the machine is off. I tried
temporarily removing the battery to reset the BIOS and the result was
that the time reverted back to several years earlier however the seconds
were ticking normally. I entered the correct time/date and restarted the
computer and all was well. I turned it off for a while and then powered
it on again and entered the BIOS setup. The time was stuck again and the
seconds were not advancing. It seems to me that no matter what condition
the Windows operating system is in, it shouldn't be able to interrupt
the time function when the set is powered off. I tried updating the BIOS
to a newer version but that just crashed the machine. (Fortunately it
didn't brick the machine). I found other users complaining about the
frozen clock issue but as yet have found no solution. Here is an example
of a similar complaint.
https://www.dell.com/community/Laptops-General-Read-Only/date-time/td-p/4122751

Anyone have any ideas what is causing this suspension of time?

Thanks for your reply.

--
David Farber
Los Osos, CA

[John Robertson]

Something is wrong with the on-board RTC - Real Time Clock.

You need to identify that chip or module and first see if its crystal is
damaged.

If the device is a Dallas TimeKeeper then you can simply plug in another
one, if soldered into the board then you have to decide how much it is
worth to fix...

John :-#)#

--
(Please post followups or tech inquiries to the USENET newsgroup)
John's Jukes Ltd.
MOVED to #7 - 3979 Marine Way, Burnaby, BC, Canada V5J 5E3
(604)872-5757 (Pinballs, Jukes, Video Games)
www.flippers.com
"Old pinballers never die, they just flip out."

[David Farber]

Thanks John,

It is worth it to me to know how much trouble it is and then decide if
it's worth fixing. I'll let you know what I find.

[Mike]

I had a laptop with a bad cmos battery.
Had weird symptoms related to timekeeping, but the cmos settings were
retained.
The voltage was low, so I changed the rechargeable battery.
That fixed it.

[David Farber]

Hi Mike,

The CR2032 battery is not rechargeable and it is not measuring below 3.0
volts. There were several other people reporting this Dell frozen time
problem and replacing the battery did not fix the issue.

[Andy Burns]

John Robertson wrote:

> If the device is a Dallas TimeKeeper then you can simply plug in another one

A DS1287 is probably thicker than most laptops :-)

[Adrian Caspersz]

On 25/03/2019 21:59, John Robertson wrote:
> On 2019/03/25 10:28 a.m., David Farber wrote:
>> This Dell Inspiron N5110, which has Windows 7 Home Premium installed,

>

> Something is wrong with the on-board RTC - Real Time Clock.
>
> You need to identify that chip or module and first see if its crystal is
> damaged.
>
> If the device is a Dallas TimeKeeper then you can simply plug in another
> one, if soldered into the board then you have to decide how much it is
> worth to fix...
>
> John :-#)#
>

Schematic available here

[1]
http://kythuatphancung.vn/download/dell-inspiron-n5110-wistron-queen-15-laptop-schematics.html

The RTC is part of the 'Cougar Point Platform Controller Hub'
[2] https://www.mouser.com/pdfdocs/6chipsetc200chipsetdatasheet.pdf

Clearing CMOS is apparently done by a jumper (G2101) RTCRST# (page 21 of
the schematic [1]. Power off then momentary short (p.163 of [2])

--
Adrian C

[Look165]

The main cause, as explained, is the battery (CR2032).
You need to change it.
It causes also BSOD.

[John Robertson]

On 2019/03/26 5:07 a.m., Look165 wrote:
> The main cause, as explained, is the battery (CR2032).
> You need to change it.
> It causes also BSOD.

The OP did mention that he tested the CMOS battery and it read above 3.0VDC.

Now it could be the battery holder that is defective. You need to check
that Vbb is getting to the timekeeper - and someone kindly posted a ink
to the schematics so the OP can check that out now more easily.

John :-#)#

[David Farber]

On 3/25/2019 3:36 PM, KenW wrote:

> I found a free program Neutron Time Set that I run from startup that
> sets the time for me.
>
>
> KenW
>
Hi Ken,

Thanks for the workaround idea. I may have to resort to using it!

[David Farber]

Hi Adrian,

That schematic and datasheet will be very helpful. Thanks!

[David Farber]

On 3/26/2019 8:58 AM, John Robertson wrote:
> On 2019/03/26 5:07 a.m., Look165 wrote:
>> The main cause, as explained, is the battery (CR2032).
>> You need to change it.
>> It causes also BSOD.
>
> The OP did mention that he tested the CMOS battery and it read above
> 3.0VDC.
>
> Now it could be the battery holder that is defective. You need to check
> that Vbb is getting to the timekeeper - and someone kindly posted a ink
> to the schematics so the OP can check that out now more easily.
>
> John :-#)#
>
>>
>
>

Hi John,

I was testing the battery voltage where the battery holder's terminals
are soldered to the board.

Now that I have the schematic, it should be fairly simple (famous last
words, right?) to track down the fault.

[John Robertson]

Nothing better than fixing stuff IMHO.

We generate enough garbage that keeping something going that is still
useful is worth the effort, puls the brain cells need exercise!

John ;-#)#

[Adrian Caspersz]

On 26/03/2019 17:15, John Robertson wrote:
> On 2019/03/26 10:12 a.m., David Farber wrote:

>> Now that I have the schematic, it should be fairly simple (famous last
>> words, right?) to track down the fault.
>>

> Nothing better than fixing stuff IMHO.
>
> We generate enough garbage that keeping something going that is still
> useful is worth the effort, puls the brain cells need exercise!

Quite :)

The reason why I pipped in, was that I was researching a real time clock
/ NVRAM fix to a 27-year old Sun X terminal [1], and came across a
description of a 'kick start' procedure to initialise the installation a
new clock module (a 28pin DIL chip like the popular DALLAS type).
However, I just needed a new battery fitted to my old one, and set about
mine with a hacksaw like this ...

http://www.glitchwrks.com/2017/08/01/gw-48t02-1

[1] For the hell of it, and a break from coding with much faster RPi
things, I've been wandering down retro memory lane putting life back
into an old discarded Sun Sparc Classic X. (removing the X terminal
nature of the beast and installing a 1GB SCSI HDD, 72-pin parity SIMM,
NetBSD 8.0 etc).

It doesn't really have a purpose (unit lacks USB, CPU runs at 50MHz) but
the retro experience has reignited some brain cells - so it's kind of
therapeutic. In twenty-thirty years time, I'll probably have the same
fun with old Core2Duo driven things like what I am using here.

For a healthy mind, I heartily recommend this time travelling retro
hobby. Sure beats sudoku and eating lots of fish ...

--
Adrian C

[David Farber]

I cosign on salvaging electronics and exercising the brain cells.

By the way, how is it the computer keeps perfect time once the time is
set and the power remains on? Shouldn't the time shown in the BIOS setup
screen begin to advance too once it's powered on?

[Look165]

BIOS settings are kept by the battery, not only time and date.
IF time slips or BSOD, it means the battery is dead or holder.
If not, Flash EEPROM is dying.
Anyway, if your PC can work, try HW32INFO.

David Farber a écrit le 25/03/2019 à 18:28 :

[David Farber]

On 3/26/2019 2:54 PM, Look165 wrote:
> BIOS settings are kept by the battery, not only time and date.
> IF time slips or BSOD, it means the battery is dead or holder.
> If not, Flash EEPROM is dying.
> Anyway, if your PC can work, try HW32INFO.
>

The computer does not BSOD since I used a restore point to correct that
issue. The battery is not dead. I measured it mounted in the holder at
the holder's terminals. The PC works fine except for the clock. Is there
some particular piece of information you want me to find with HW32info?

[whit3rd]

On Tuesday, March 26, 2019 at 10:12:02 AM UTC-7, David Farber wrote:

> I was testing the battery voltage where the battery holder's terminals
> are soldered to the board.

Good. And, the usual self-test can/should exercise the chip that
keeps the timer running. It won't, necessarily, test the time accuracy,
and a typical such system uses a quartz crystal with a couple of
small capacitors. The crystal could be fractured, or (worse) one
of the capacitors could be shorted. Tiny capacitors, only gonna
be able to find 'em because they're next to the rock.

[whit3rd]

On Monday, March 25, 2019 at 2:59:29 PM UTC-7, John Robertson wrote:
> On 2019/03/25 10:28 a.m., David Farber wrote:

> > This Dell Inspiron N5110... The system time does not advance

> > when the unit is turned off.

> Something is wrong with the on-board RTC - Real Time Clock.
>
> You need to identify that chip or module and first see if its crystal is
> damaged.

Another poster has suggested a circuit diagram
<http://kythuatphancung.vn/uploads/download/22163_DELL_N5110.pdf>

and the crystal would be X2101 and its associated components, R2101, C2101, C2102,
depicted on 'sheet 21 of 108' (see the label on lower left page corners) and connected
to pins A20 and C20 of the 'Cougar Point' big integrated circuit.

[David Farber]

On page 99 of the schematic you can see the CMOS battery output labeled
as RTC_AUX_S5. I did a text search for that and it took me to page 27
which shows VBACKUP. There are two partial circuits there. One appears
to be for the CMOS battery backup and the other one, 3D3V_AUX_KBC,
appears to be the backup power supplied when the unit is turned on. Both
of these circuits point to EC GPIO72. Regarding the RTC_AUX_S5 diagram,
there's an orange ellipse around the specifications for a resistor. It
says 10mW 0R0402-PAD-2-GP. Might that be a fusible link? Also, I see the
words "stuff" and "un-stuff" appearing frequently in the schematic. Are
those synonyms for install and remove?

[Adrian Caspersz]

On 26/03/2019 21:21, David Farber wrote:

>
> I cosign on salvaging electronics and exercising the brain cells.
>
> By the way, how is it the computer keeps perfect time once the time is
> set and the power remains on? Shouldn't the time shown in the BIOS setup
> screen begin to advance too once it's powered on?
>

The RTC is only read once on startup of the operating system, which then
maintains the increment of it's own internal counter. There is no
'write-back' if current OS time settings are left alone by the user.

The BIOS time itself not incrementing is the fault, I reckon the chipset
has got itself into a funny state and needs a reset which you haven't
yet done. I'd try finding the G2101 link I mentioned.

Interestingly, the ACER ASPIRE V5 471 has the same labelled jumper,
similar circuit (same original manufacturer) - and googling that
"G2101", it looks a known reset method for locked BIOS etc. That Acer
motherboard visually looks different, but the G2101 triangular pads were
located under the DIMM sockets. Maybe Dell is in a similar location.


--
Adrian C

[Adrian Caspersz]

On 27/03/2019 17:36, Adrian Caspersz wrote:

> Interestingly, the ACER ASPIRE V5 471 has the same labelled jumper,
> similar circuit (same original manufacturer) - and googling that
> "G2101", it looks a known reset method for locked BIOS etc. That Acer
> motherboard visually looks different, but the G2101 triangular pads were
> located under the DIMM sockets. Maybe Dell is in a similar location.
>
>

http://1.bp.blogspot.com/-CNHXkSjj8VE/UNV1-52T3uI/AAAAAAAAAR0/CpxDrJtoCjc/s1600/Dell+5110.JPG

G2101 is down to the right of G2201 as indicated. You don't want to mess
with G2201, that's something else.

--
Adrian C

[David Farber]

Hi Adrian,

Your above link to the picture of the G2201 location is correct. Please
note that it's on the opposite side of the motherboard where the CMOS
battery lives. After performing the reset, I went into the BIOS setup
menu and set the time and date. Since then, the displayed time is now
accurate even after being powered off and then powered back on again.
This just goes to show you that removing the CMOS battery does not
necessarily reset whatever it was that needed resetting. (At least for
this computer)

Thanks to everyone for all your fantastic support!!!